en.Wedoany.com Reported - Amazon Web Services Inc. announced today that vector search for DynamoDB, its high-availability NoSQL key-value and document database, is now generally available.

Since its launch in 2012, DynamoDB has undergone numerous iterations and is now positioned as the database of choice for developers who need predictable performance without operational overhead. The service is suited for serverless web and mobile applications, gaming, ad tech, IoT, retail, and business scenarios requiring high throughput and low-latency access.

The newly available vector search capability delivers single-digit millisecond latency and 99% recall, designed to support data at any scale, including trillions of vectors. Developers do not need to provision, patch, or manage any servers.

This update enables DynamoDB to power large-scale applications requiring semantic retrieval, covering use cases such as AI applications and agent memory, retrieval-augmented generation (RAG), recommendation engines, personalization, and anomaly detection.

The core role of vector databases in AI development is to store content as high-dimensional data called "embeddings," enabling fast similarity search based on meaning rather than keyword matching. Many AI systems use vector databases as long-term memory and process large volumes of unstructured data to build context for large language models and agents, improving response accuracy and preventing erroneous outputs such as hallucinations.

Many developers were already retrieving business data on DynamoDB. With the addition of vector search, developers can now migrate vector retrieval capabilities to the same service that also hosts their business data, sharing the same managed serverless infrastructure and pay-per-request pricing. Previously, developers had to run two separate systems for primary data storage and vector search.

Currently, native vector search capabilities are available across multiple of the company's specialized database services. Key vector services include S3 Vectors, a native storage, indexing, and sub-second similarity query engine for cloud object storage, and OpenSearch Service, which offers fully managed vector search with a serverless vector engine for billion-scale datasets.